About Me

My photo
Australian philosopher, literary critic, legal scholar, and professional writer. Based in Newcastle, NSW. My latest books are THE TYRANNY OF OPINION: CONFORMITY AND THE FUTURE OF LIBERALISM (2019); AT THE DAWN OF A GREAT TRANSITION: THE QUESTION OF RADICAL ENHANCEMENT (2021); and HOW WE BECAME POST-LIBERAL: THE RISE AND FALL OF TOLERATION (2024).

Monday, September 05, 2011

Not so odd - Jerry Coyne on Joel Marks

Brother Coyne - sorry, Jerry, I couldn't resist! - has a piece on Joel Marks in which he describes the latter's contribution to Opinionator as "a very odd column". I hope that readers of this blog, where we often discuss issues in metaethics, will understand why I don't think it odd at all. Marks has embraced a perfectly respectable position in philosophical metaethics: what looks like a form of moral error theory, closely akin to the abolitionist position of Richard Garner.

There's some confusion to be cleared up here, and it's much the same confusion that is often shown by Sam Harris. When we talk about morality being "objective" we are not interested in whether people's moral choices have certain properties that are independent of our desires - of course they do. Thus, it may be that a particular choice will lead to the suffering of certain people or certain other sentient animals. Another choice may lead to certain pleasures or satisfactions for those affected, while causing them no pain or other harm. Nothing is controversial about this, but pointing it out won't take us far towards saying that there are objective moral requirements in the sense debated by metaethicists.

(Well, it might; it might lead us to moral naturalism. But moral naturalism gives a rather deflationary account of moral objectivity; and in any event it has to be argued for, which is easier said than done.)

We can get closer to what an objective moral requirement would be when we think - by contrast - about ordinary value judgments. Let's take the good (!) old car example again. When I say, "This is a good car," I mean something like, "This car has the properties that we desire in a car." If I say, "This car (my friend's Porsche, say) is a better car than mine," I mean that it has those properties to a greater degree (perhaps using some sort of weighting system to add up the relevant properties).

My friend's (hi! I know you're reading this!) Porsche really does have certain characteristics, as does my Honda. But am I inescapably bound to prefer her car to mine irrespective of my particular desires? No, I am not. The "goodness" of the car is parasitic not only on properties (such as its mass, various performance characteristic, etc.) that will be the same whatever our desires may be, but also on what our desires actually are. Someone whose desires differ from those of the majority cannot be forced, on pain of being just wrong about something, to prefer one car to the other. There is an important sense in which car-goodness is not an objective property. It is not Out There, independent of how we feel about the characteristics of cars. It can't prescribe how I act, for example in purchasing one car rather than another ... on pain of being just wrong if I act in some other way.

With most value judgments, there's not a lot of intellectual resistance to this analysis. But the situation tends to change with moral judgments. If someone says, "Torturing babies is wrong," she is likely to mean something much stronger than, "Torturing babies is painful for them." The latter statement is doubtless true, and it is true independently of my desires in the matter. But a sadist who goes ahead and tortures babies may be making no mistake about anything in the world (though likewise we make no mistake when we judge the sadist to be "evil").

There is a pressure, perhaps built into our language itself, to regard the "moral wrongness" of torturing babies as an objective property, in the sense that car-goodness is not: to regard it as a property out there in the world independent of our desire for babies not to suffer pain.

But of course there is no such "objective" property. The pain exists. Our desire that the pain not happen exists. The mismatch between the pain and our desires exists. It is rational for us not to torture babies. It is rational for us to attempt to prevent others from doing so, and to enact laws that deter others from torturing babies. All of that's fine. And yet, there is this pressure (where does it come from? I'm sure some of my opponents will attribute it to God ... but that has all sorts of problems) to want something more, and to believe that this "something more" actually exists.

We seem to want this further property of moral wrongness that is independent of desires. I.e., we want this propert to exist. We seem to find it difficult to admit that there is no such property.

We seem to want objective prescriptivity to exist, even though objective prescriptivity would be a very queer thing indeed, as moral error theorists point out. Or at least many of us seem to want this, and to be convinced that there is such a thing. And yet, no one has ever seen such a thing, and it is hard to understand how the concept is even coherent.
Joel Marks has reached a point in his life where he no longer believes in objective prescriptivity, in inescapably binding standards of conduct, or in morally evaluative properties ("moral goodness", "moral forbiddenness", etc.) that are independent of our desires. For Marks, the "goodness" of choices and actions had better be no more metaphysically rich than the "goodness" of a car: a kind of summation of whether they have the metaphysically ordinary properties that "we" (and this "we" is always, in practice, at least partly a fiction) want in choices and actions of that type.

It might be said with some truth that Marks is leaving morality behind. That doesn't mean that he is leaving goodness behind. He might still make choices and carry out actions that we regard as "good" - perhaps because they conform to our inherited standards, or because, in any event, they have the properties that we want such choices and actions to have. But he is living without what is sometimes called the moral overlay, the illusion of desire-independent objectively prescriptive properties out there in the world.

It that sense, Marks is, arguably living beyond morality, though he need not necessarily be living beyond making value judgments (as he describes the situation he even seems to be doing that, but I think that's a further step).

As I said at the start, I don't find any of this odd. It's a difficult concept to convey at less than book length to people who are not already familiar with it. But it's a genuine issue in metaethics, and the problems it raises are quite familiar to people working within the field. Marks may still live much as he did before, and we may still make value judgments about him - he can't escape being judged by others. We may judge him to be a "good" or "virtuous" man, if all we mean by that is that he has the dispositions of character that we want in a man. For all that, he's thinking about his choices in a way that most people don't when they imagine that there are inescapably binding moral requirements.

I'm not sure I've conveyed this well. It really is difficult to tease out briefly, and I don't think I've done very well in oral presentations in the past or even, necessarily, in previous blog posts like this one. But something really has changed for Marks. He's abandoned objective ethics - not the belief that there are properties of our choices that exist independently of our desires, such as whether or not a choice, when acted on, will produce pain, but, rather, the belief in objectively prescriptive properties such as the property of being inescapably forbidden.

The more important question, perhaps, is whether this change - this movement beyond morality as it's commonly conceived - really matters. Marks thinks it doesn't. I tend to agree. Actually, I think it helps us to see things more clearly, and to consider what we think moral systems are really for.

That question will not have a straightforward, literal answer, since moral systems were not consciously designed, so they are not literally "for" anything. Still, the question makes a kind of sense. We can ask how moral systems actually function, and we can ask what we actually want in a moral system, if we still want such a thing at all. This then starts to give us a basis to assess - and, where appropriate, even debunk - the traditional morality that we've inherited.

How much of our traditional morality will we still approve of when we've subjected it to radical scrutiny? Maybe some it has to go.


James Sweet said...

The issue I have with this analysis is that it leaves out the idea that while desires may differ, often leading to different evaluations of "goodness", there is enough of a universality to some desires to make some evaluations (whether they be of the moral or automotive variety) so close to "objective" that we would consider it pathological for someone to disagree.

I've tried to extend this into the car metaphor in the past, but I am now thinking a better might be food. Surely, there is no objective answer to the question, "Which tastes better, chocolate ice cream or vanilla?" But what about the question, "Which tastes better, chocolate ice cream, or a hunk of cardboard?" You might still argue that the answer is not really "objective" as such... but if someone answered that cardboard tasted better, we would not shrug our shoulders and say, "I guess your desires are different from mine" -- we would label it pica and consider it a pathology. (I'm leaving out here for the moment the technical possibility of some sort of allergy that would make the ice cream fatal and the cardboard merely unpleasant -- let's not overextend the analogy)

By the same token, we might well tolerate different answers to the question, "If you cheated on your spouse with a one night stand, and you feel awful and are unlikely to ever do it again, are you obligated to tell him/her?" But someone who gives the "wrong" answer to the question, "Is it fun to torture babies?" we will surely label a sociopath and categorize that as a pathology.

Does this get us all the way to "objectivity" for at least some moral questions? I know a dedicated moral error theorist such as yourself will say no, and I suppose I have to mostly agree (though maybe we can say it is "objective" if we define morality in relation to the evolved traits of H. sapiens, but believe me that I recognize how dicey this proposition is). But it is so close to objective that I'm not really sure I see where the practical difference is -- and you touch on this in your last few paragraphs here.

It's somewhat analogous to your position on free will: That libertarian free will clearly flies in the face of physical determinism, but that the more interesting questions revolve around the compatibilist position. By the same token, true moral realism clearly doesn't work with a materialist universe, but maybe this sort of purity is not where the interesting questions are.

At one point I considered labeling my position "moral quasi-realism," then I found out that term has already been taken -- but it may be pretty close to what I am talking about. I need to read more about it...

James Sweet said...

Let me give an example that I think will illustrate how it is the universality of certain desires that makes us class them as having to do with morality:

I think so-called New Country music sucks. Like, really sucks. In fact, I might even go so far as to say that in some sense, it objectively sucks, and I think I could make a case for why it is not just music I find unpleasant, but that it is truly "bad" music and everybody ought to think so.

Stealing a pack of chewing gum from the supermarket also sucks. But if I were given the choice of living in a world where nobody ever stole a pack of gum from the supermarket anymore vs. living in a world with no New Country music, well... let's just say Toby Keith might have a lucrative fallback career running a Dentine smuggling ring.

Yet despite this, I think I could make a strong case that stealing a pack of gum is immoral, while I would have a lot of difficulty saying that New Country music is immoral.

If morality is nothing more than a desire misclassified, then why should that be the case?

Straight-up moral error theory, while it is probably technically correct, has nothing to say about this last question, at least as far as I can tell. It seems to be leaving a lot out.

Marshall said...

I wouldn't like to say that Marks "is leaving morality behind." It seems to me that moral thought does have a distinct quality, a felt compulsion. If we are thinking about some act of cheating (financial or marital), talk about "cheating" brings in considerations other than "will it be fun for me", "can I get away with it" and so on. It isn't just "personal desire" like choosing Ranch or Italian dressing. Consequential "good/bad" considerations are still alive for Marks, it's just that he sees that those considerations arise within himself rather than being imposed from outside. This isn't "amorality" in any normal sense of that word.

Marks gets to choose his own car rather than accept one his father bought for him. Either way he's rattling around town in the same loveable old Honda, but this way he is pushed towards thinking about "what car-goodness is, what it is for", to take responsibility for his motivations as well as his actions. I claim that difference, which you seem to sweep aside as a "doesn't really matter", is a big deal. First order sameness, second order big difference.

(If I read her correctly, Jean thinks this transition to internal moral values leads to reduced commitment, yes Jean? Whereas I think it enhances commitment. (Off topic,
more on Roosters here

suffering of certain people or certain other non-sentient animals.

I think you mean 'sapient' rather than 'sentient'. The Buddhist category of 'sentient beings', which is large and ill-defined, seems to me a better fit than 'persons'. As in, "although sentient beings are numberless, I vow to be concerned with each one of them." Non-human animals are clearly sentient although they are not sapient, or at least not as sapient as we are.

Anonymous said...

I think you did convey this well. I found Jerry's criticisms of Marks odd, considering that both of them seem to believe the same thing - that moral laws do not exist in their own right. It's good to see I'm not the only one.

Tim Martin

Russell Blackford said...

One correction made - I wrote "non-sentient" at one point when I meant "sentient". Fixed.

Russell Blackford said...

James, I agree with your first paragraph. I think that's exactly right. It's more a point for my next post, when I tackle Jean Kazez's criticisms of Marks, but it's a good point and in fact it's important to my thinking on all this. As I've said now and then, our moral systems are not just arbitrary.

Still, I don't think that's a reason why Marks has to go on thinking in moral terms rather than in the terms that he describes.

I also agree that straight-up error theory doesn't in itself tell us how to live our lives. But it's not meant to replace all thinking on how we live our lives. At most, it can clarify some things that might then help us with the rest of our reasoning about such issues.

And thanks Marshall about the animals - I think I just meant to write "sentient", rather than "non-sentient" ... maybe I meant to write "non-human sentient animals" or something. Anyway, fixed now as "sentient".

Svlad Cjelli said...

"It can't prescribe how I act, for example in purchasing one car rather than another ... on pain of being just wrong if I act in some other way."

Nitpick: You can be wrong about the properties of the car.

If a person tries to serve "human wellbeing" by killing others so that they go to heaven, the reality of this heaven factors into whether he is just wrong or not.

Richard Wein said...

I think it can help to concentrate on moral "oughts" rather than moral "is's" like "X is morally wrong". The sense of prescriptiveness or obligation is clearer when we talk about "oughts". It may also be useful to think in terms of "reasons for action". Anwyway, for what it's worth, here's one alternative way of looking at this.

To say that one "ought" to do something is to say that one has good reasons to do it. Knowing that someone wants to catch a particular train, I might say, "You ought to leave now". In this case I'm imputing no obligation to leave now. I'm just giving practical advice. In a practical case like this, the good reasons are ultimately rooted in the subject's own desires, combined with facts about the world, such as how long it will take to reach the station. He ought to leave now because that's the best way to fulfill his desires.

"Desire" here should be understood in a very broad sense, to include anything the subject cares about, not just selfish wants. If someone says that he wants to promote the general well-being, and I reply that he ought to give to charity, I can be just giving him advice on how to achieve his desire. I'm not necessarily imputing any obligation on him to do so.

But _moral_ oughts involve a different type of reason. When we say that someone has a moral obligation to do something, we mean that he has a good reason to do it regardless of whether it will fulfill his desires. For example, "You ought to give to charity, even if you have no interest in doing so". But what good reason can the subject have to do something that doesn't in anyway fulfill his desires? Whatever his most basic moral claim may be (e.g. one ought to promote the general well-being, or to follow God's commands) all the moral realist can do is to insist that the subject has good reason to act on it, without being able to give any justification or even to explain what kind of reason he is talking about.

My case against moral properties (incuding moral reasons for action) is an inference to the best explanation based on the following observations:
1. There's no evidence for their existence, beyond the fact that most people believe in them.
2. No one can derive a moral fact without starting from another moral fact. (Can't derive an "ought" from an "is".)
3. Moral realists cannot give a successful explanation of the meaning of a moral term without employing another moral term.
4. We don't need moral facts to explain any observation. We can explain moral discourse and behaviour as a consequence of people's moral beliefs. The premise that these beliefs correspond to facts is superfluous. We can (roughly) explain people's moral beliefs as the result of evolutionary and social factors.

Russell Blackford said...

Svlad, everyone agrees that you can be wrong about the properties of the car. You might think it has a fuel consumption of 6 litres per hundred kilometres when it actually burns up 12 litres per hundred kilometres. That could make a big difference to how you judge it and whether you buy it.

Getting the naturalistic properties right is important (which is one reason I'm puzzled why Jean seems to think that there's no room for rational discussion on this picture - it seems to me that there will often be plenty of room, whether we're talking about ordinary value judgments or judgments about choices and actions).

Svlad Cjelli said...

"2. No one can derive a moral fact without starting from another moral fact. (Can't derive an "ought" from an "is".)

I think it's arguable whether "James wants more cake" is a moral fact or just a fact-fact. Either way, it informs what he ought to be doing.

Russell Blackford said...

Actually, I agree that you don't need moral facts to derive "oughts". All you need is a goal or a desire or a want, or something like that, plus information about how to achieve it. But note that you have not derived an "ought" that is desire-independent (using "desire" broadly). That was what we set out to do. That is the essence what is meant by the claim that morality is objective.

Hume never denied that it was possible to derive oughts that are desire-dependent. No one really denies that (although there are people who'll quibble about it, e.g. by saying that an analysis in terms of desires ends up being trivial, or something of the sort).

Error theorists typically claim that people think of moral claims as being about properties of "objective moral wrongness", "objective forbiddenness", etc., that are "out there" independent of our desires (and of institutions such as etiquette, law, social mores, etc.) - and that that is what moral claims, or at least a large class of them, mean.

But because no such properties exist, error theorists say, moral discourse is full of statements that are, strictly speaking, false.

Again, a moral error theorist does not have to claim that moral judgments have no true content at all and no non-cognitive content. They may have fairly complicated content.

As I've said in the past, I'm not so much concerned about whether the semantic part of this thesis is true. It probably (I think) does have some truth to it, but perhaps it needs to be qualified carefully. But regardless how much the language is committed to the existence of these strange desire-independent prescriptive properties, i.e. expressions such as "morally wrong" just do have that as part of their meaning, a lot of people (including a lot of philosophers) seem to think these properties exist.

Richard Wein said...

Russell wrote: 'Actually, I agree that you don't need moral facts to derive "oughts".'

I'm not sure if this was in any way a response to my comment. If so please note my distinction between moral and non-moral oughts. I agree that non-moral oughts can be derived from other non-moral facts. It's moral oughts which I say can't be.

A given utterance may combine both moral and non-moral oughts.

Tom Clark said...

"But note that you have not derived an "ought" that is desire-independent (using "desire" broadly). That was what we set out to do. That is the essence what is meant by the claim that morality is objective."

I'm puzzled as to why it was ever thought possible that moral oughts or more broadly anything normative could exist independently of desires, needs, goals, etc. Perhaps the whole tradition of moral realism/objectivity is kind of a hold-over from theism, where's there's an external source of moral law. But as others have pointed out, there's a huge difference between anti-realism and nihilism, since broad agreements on norms flow from our shared human nature.

Richard Wein said...

I think it's confusing to talk about the dependence of oughts on desires without specifying _whose_ desires we are talking about.

A practical (non-moral) ought--as I described it above--is rooted in the desires of the _subject_ of the ought, i.e. the person who ought to do something. It says how the subject can best achieve his desires. It doesn't say that he has any obligation to do it.

An ought that's allegedly rooted in desires of _other_ people is likely to be a moral ought. To claim that you ought to do something because it's what your parents desire (even though it's contrary to the fulfillment of your own desires) is to claim that you have an obligation to do it.

At the risk of complicating matters, perhaps I should mention another possible type of non-moral ought: an ought of human convention, e.g. a legal ought. A speaker might say: "Legally you ought to do X [or are legally obliged to do it] but as the law is a bad one you have no moral obligation to follow it, and therefore no moral obligation to do X." Arguably, this sort of non-moral ought does assert an obligation of sorts, and such an obligation exists in some sense, but it's not an _actual_ obligation. It's just an obligation "according to the law". Is this sort of ought rooted in any desire? Well, I suppose you could say that the convention wouldn't exist unless some people desired it, but that's a rather different type of desire-dependence.